如果因变量的残差方差齐，不同预测值对应的残差应大致相同。即图中各点均匀

If the residual variance of the dependent variable is uniform, the residuals corresponding to different predicted values ​​should be approximately the same. That is, the points in the figure are evenly distributed, and no special distribution shape appears. <br>If the residual distribution is not uniform, forming a funnel or fan, then the variance is uneven. In the scatter plot of the predicted values ​​and standardized residuals of the dependent variables within each group, the points are evenly distributed and have no special shape. Therefore, it can be considered that the residual variance of the dependent variable within each group is uniform.

If the residual variances of the variables are aligned, the residuals corresponding to the different predicted values should be approximately the same. That is, the points in the figure are evenly distributed and no special distribution shapes appear.<br>If the residuals are unevenly distributed to form funnels or fan shapes, the variances are uneven. In the scattering chart of the predictors and standardized residuals of the cause variables in each group, the points are evenly distributed and there are no special shapes. It can therefore be argued that the residual variances of the cause variables within each group are aligned.

If the residuals of the dependent variables are homogeneous, the residuals corresponding to different predicted values should be roughly the same. That is to say, each point in the graph is evenly distributed, and no special distribution shape will appear.<br>If the distribution of residuals is uneven, forming a funnel or fan, then the variance is uneven. In the scatter plot of the predicted values of the dependent variables and the standardized residuals in each group, the points were evenly distributed without special shape. Therefore, it can be considered that the residual variance of the dependent variable in each group is homogeneous.<br>